DESCRIPTION (Adapted from applicants abstract): Despite advances in chemotherapy, acute leukemia remains the primary disease related cause of death for children in the U.S., affecting ~2,500 children annually; 45,000 adults will die in 1989 of leukemia or lymphoma. While most patients can achieve remission, leukemic relapse often proves fatal. Allogeneic bone marrow transplantation (BMT) is potentially curative but many still relapse after BMT. The "graft-versus-leukemia" (GvL) effect strongly suggests that immune cells within the BMT may help prevent recurrence. Since 1982, this project has analyzed destruction of malignant hematopoietic cells by human lymphocytes to characterize effector/target cell interactions and to identify cell types that recognize human leukemia. Murine immunotherapy data and clinical GvL effects suggest these lymphocyte-leukemia interactions may prospectively provide antileukemic therapy. This proposal describes five years of integrated in vitro research that focuses on cellular and molecular leukemia recognition mechanisms. Each aim will expand and clarify recent preliminary data. We propose to analyze: 1) The frequency, HLA regulation and molecular specificity of T lymphocytes clones that specifically differentiate between allogeneic leukemic blasts and remission cells from the same patient. Some cloned T cells primed to allogeneic leukemic which appear leukemia specific will be studied. This aim includes a major effort to generate "GvL" T cell clones directed to Ph+ chromosome controlled BCR-ABL protein sequences that may recognize Ph+ leukemia cells. 2) The frequency, specificity and HLA regulation of T lymphocytes able to recognize autologous leukemia cells. In vitro expansion with Interleukin-2 (IL-2) of cells from selected leukemia patients has generated clones with specificity reactivity to autologous leukemia. We will determine their frequency, HLA control, and molecular target, as in Aim #1. 3) The leukemia-reactivity and specificity of gamma/delta clones that mediate non-MHC restricted cytotoxicity (NRC). Both NK cells and T cells (especially gamma/delta cells) can mediate NRC to leukemia cells and could account for the in vivo "GvL" effect. Analysis of >800 NRC clones from a single donor showed a unique specificity pattern. Recognition mechanisms of these gamma/delta clones will be characterized. 4) The role of "novel" and "public" MHC determinants in restricted leukemia recognition. HLA deficient mutant B cell lines will enable analysis of the role of specific MHC molecules in recognition by leukemia reactive T cell clones. 5) Test in vitro the in vivo antileukemic potential for leukemia-reactive lymphocytes. In vitro analysis of these antileukemic reactions will provide data testing their therapeutic potential and help direct better in vivo application of the antileukemia reactivities that are generated.